W3C home > Mailing lists > Public > public-xml-processing-model-wg@w3.org > July 2006

p:pipeline

From: Norman Walsh <Norman.Walsh@Sun.COM>
Date: Wed, 19 Jul 2006 16:04:40 -0400
To: public-xml-processing-model-wg@w3.org
Message-ID: <87wta9uzs7.fsf@nwalsh.com>
In this and the following few messages, I'm going to try to summarize
where I think we are. I'm trying (perhaps foolishly) to do this with
a slightly more abstract and compact syntax.

A pipeline looks like this:

  pipeline := ($name?, input*, output*, {body}+)

That is, it has an optional name, zero or more inputs, zero or more
outputs, and a body. The body is basically "step" but will actually be
(step|for-each|choose|...)

An input looks like this:

  input := ($name, $label, $href?, {any content})

The name/label pair form the external/internal binding for the stream.
If the API doesn't define the input, then it can be read from a URI or
taken directly from content.

An output looks like this:

  output := ($name, $from)

The name/from pair form the external/internal binding for the stream.
The from must point to the label of some in-scope (a concept not yet
clearly defined) output (e.g., the with-output on a step).

                                        Be seeing you,
                                          norm

-- 
Norman Walsh
XML Standards Architect
Sun Microsystems, Inc.

Received on Wednesday, 19 July 2006 20:04:46 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 8 January 2008 14:21:48 GMT